This paper discusses technology trends and foresees their impact on college teachers. "Teaching" institutions have perhaps a greater obligation to adopt teaching technologies than do research institutions that boast primarily of library and laboratory facilities. The Internet changes how we teach because of its potential to provide any kind of information (including lectures and help desks) on demand. The electronic classroom magnifies our presentation capabilities, making these presentations both more intense and potentially distributed. Our challenge is to design instructional systems that will allow instructors to interact with a student at the other end of a wire as effectively as if the student were at the other end of a log.

The growth of the Internet has roughly paralleled the growth of compressed video for distance learning. In Louisiana a state agency has been working for five years setting up standards for distance learning classrooms, configuring a statewide control center, and building a network among colleges and universities. They are well behind the infrastructure developed in the state of Texas. LSU-Shreveport has been receiving interactive video courses from LSU for three years. Although the number of distance learning classrooms in the state is set to double in the next two years, the focus on room conferencing, ISDN technology, is already obsolete. The next generation of interactive video is out of the development labs and into commercial application. Unlike the ISDN approach, the next generation technology will be integrated with the next generation Internet and will focus on many-point interaction using H.323 (IP-based) communication protocols. This revolution in transmission capability will take time to filter to non-urban areas, just as digital TV is here but not everywhere. However, the rule applies that hardware is always more capable than software, and that software is always more flexible than our imagination on how to use it (we are still thinking of novel ways to use spreadsheet programs). While it is always possible to swamp the CPU or the network with brute force calculations, their performance is growing so rapidly that even problems for which there are no clever algorithms are now computable (witness the frailty of low strength encryption). As the performance of our communication systems rise and the costs drop, teachers must be continually re-evaluating the cost/benefit ratio of instructional technology. This means that teachers must be applying more imagination to the teaching/learning dynamic, revising the design of their learning experiences, and experimenting with previously impractical approaches to educating students.

Synchronous compressed video between classrooms is one distance learning strategy, synchronous point-to-point video conferencing with white board is another, and asynchronous email/web page interaction is a third. It doesn't take much imagination to ascertain which alternatives are temporary kludges forced by limited budgets and technologies. For the purposes of effective instructional interaction, we want face-to-face relationships supplemented by the ability to share, in real time, printed or other audiovisual information (demonstrations). Some of the time we benefit from synchronous group activities, sometimes from one-on-one tutorial situations, and sometimes from private study contexts. If travel is avoided, there is greater flexibility for participation. Therefore, our ideal would be the ability for a student or a teacher to be VIRTUALLY PRESENT at any location. This term implies that the remote person has access to the same stimuli as the local individual. We are not as far from being able to capture and distribute the most significant stimuli for education as one might think (except for hugging, which makes the virtual mode more appropriate for high education).

As LSU-Shreveport developed its electronic classrooms, we moved from overhead projectors to LCD panels and computers, to LCD projectors and Elmos, to teaching laboratories with every student at a workstation (skipping instructor controlled student workstations). The presentation aids permitted us to do computer demonstrations, to play PowerPoint productions, and to access the Internet in front of the class. These instructor-controlled tools have been implemented in all size classrooms and lecture halls, and we have even moved the equipment from room to room and from building to building on carts. As the usage has increased and the cost decreased, we now wire our rooms with network access, install a projector in the ceiling, and sometimes a permanent instructor workstation and Smartboard. However, when we moved to instructional labs that could allow students to become more active learners, permitting mini-lectures interspersed with mini-labs, we became significantly constrained by room size. The number of workstations that a room will hold is significantly less than the number of seats that it will hold, and not all rooms can be converted to teaching labs.

The conversion of lecture courses into lab courses has now been impeded because of the growth of sections required to instruct the number of students interested. The use of computing technology internally has moved us in the opposite direction of its use for distance learning–fewer students can be taught on-campus in a lab section. Our response has been to design a distance learning environment on campus. Having created two teaching laboratories within the department, we have proceeded to link them together so that we would have ONE large virtual teaching laboratory. With this accomplished, it is a small step to link INDIVIDUAL workstations at other locations on campus to the virtual lab via the campus LAN, and a slightly larger step to incorporate individual workstations off campus via the Internet. The goal is to be able to teach reasonably large sections (around 40) in a synchronous, interactive mode.

There was no excuse for linking to rooms in the same building with ISDN, so we moved directly to IP-based video conferencing technology that was not constrained by bandwidth. We have experimented extensively with several IP-based distance learning systems including ClassPoint by White Pine Software and Learning Server by DataBeam. Both of these products provide multi-point conferencing including audio, video and application collaboration (commonly called data conferencing). They both make use of inexpensive video cameras and capture cards. They provide capabilities to set up and schedule classes, register students, post class materials, and interact with students in several ways during a class session. ClassPoint allows up to 12 stations (instructor plus 11 students) to provide video at any time, while Learning Server supports only one remote video window (whoever is talking takes over the video unless controlled by the instructor). Students gain access to the classes through web browsers although both systems require additional client software to be installed for full capability.

The ability of these software products to support multi-point conferencing, as opposed to simple point-to-point conferencing, is significant because the alternative for multi-point software capability requires a much more expensive hardware MCU (Multi-point Conference Unit).

Both of the above systems however, were designed for all participants to be at desktop conferencing systems (a teacher sitting at a computer and interacting with isolated students sitting at individual computers). The VIRTUAL CLASSROOM model, however, places the teacher in front of a live class, so some adaptation of current products is required to fit our model. A room conferencing system in the primary classroom can be made to play the role of a desktop system with either the ClassPoint or Learning Server systems. An electronic whiteboard in the primary classroom is used as the application sharing display and allows for independent annotation/markup overlaid on top of the regular computer displays generated. These marked-up displays can be saved by the instructor for later recall by all.

Synchronizing the remote students (even though they were just down the hall) improves the effectiveness of the lecture for them as they can see and hear what the instructor does and they can capture the whiteboard activities. The local students also benefit from the whiteboard capture which is available on their workstations, via the network, even though they are not part of the conference. But the intensity of the instructor-controlled and paced presentation tends to preclude note-taking. The multimedia show is intended to make a deep impression on the student, but obviously this impression will not be uniformly effective. We therefore gave some thought to how to permit a student to later review what went on in the classroom.

Our solution to this problem is to implement a video server which captures all of the information that is sent to remote students (think of the server as a silent conference participant). This video repository is then made available on demand through the Internet via a web interface. Our intention is to have the the audio and video streams indexed so that a student can play back 5 minute segments of the lecture. At this writing we have not completely solved this problem, but by the end of the year we expect to have crafted a satisfactory solution (serving video lectures is practiced by several university distance learning programs). Serving video replays of the class activities is also helpful for the instructor because statistics can be generated on what parts of the class are requested for review, hence focusing attention on those parts for improvement.

The electronic classroom environment is supported by a web server and a video server that are managed through a browser interface. The video server makes it possible for all students to review class presentations at their leisure. We are planning soon to equip some workstations in the library so that students can either participate in classes or access review materials (saving a trip across campus). Our plan is then to experiment with providing access points off-campus, focusing on high schools and community colleges which already have at least T1 access to the Internet.

The foremost problem is also the simplest to solve: this application demands a lot from its servers and it utilizes fairly expensive classroom cameras. Fortunately, both these items continue to drop in price while their performance improves (an example of why teachers must keep current with the products in the marketplace). Video requires an enormous amount of storage. However, a substantial degree of compression is feasible and solves the storage problem as well as the transmission bandwidth problem. Off-the-shelf hardware and software is being used whenever possible. Browser-based clients make both the conferencing and the server access straight-forward. The instructor’s workstation makes use of the dual monitor display that is supported by Windows 98. LAN speeds insure good quality video within the campus and we may experiment with ATM to permit even greater capacity.

The most challenging technical problem is to achieve good quality real time audio and video using the Internet Protocol. The best audio and video is achieved by using integrated teleconferencing hardware/software clients such as the Intel ProShare and VCON Escort systems which do substantial real-time compression at the source. Unfortunately, these systems do not interface with many of the available software systems. While all of these claim to be H.323 standards based, the integrated systems do not provide the necessary interfaces for the use of the software packages which support multi-conferencing. This is but one example of the immaturity of the H.323 standards based components and marketplace.

VCON has announced a software release which will support multi-point conferencing and synchronous teaching/learning capabilities. This package may provide a high-quality, cost-effective solution. At the present time it is the most viable off-the-shelf system for our purposes.

By developing our distance learning skills on campus we are able to train instructors without placing students in instructional jeopardy. Any problems that occur, technical or otherwise, can be immediately addressed or circumvented. We are not asking the student to choose between instructional quality and convenience, we are demonstrating that we will add convenience without loss of quality.

The Learning Server from DataBeam will soon be packaged with Lotus’s Learning Space, an asynchronous teaching/learning product. This supports our belief is that synchronous teaching/learning is the critical ingredient of instructional quality and that asynchronous capabilities serve best as add-ons to the synchronous capabilities. We believe that as distance learning technology matures, asynchronous methods (especially web-based) will be supplanted wherever bandwidth can be obtained. We believe that most students would rather "attend" a class at an odd hour than to forego the opportunity to interact with the instructor and classmates.

Collaborative learning is being recognized as an essential modality in the growth of collaboration technology in the workplace. Student-student collaboration has been little exploited in education because the educational experience has been focused on the classroom interchange with the instructor. Educational collaboration between students has been at best a temporary exercise conducted between constantly changing collaborators with whom the student had little continuing relationship. But distance learning technology as described here can be easily scaled down to support small group collaborations. Once workstations are equipped with cameras, microphones, and high-speed communication access, at home, at work, or in the library, the potential for participating in a conference exits. Conferencing software and hardware is now available to implement virtual presence, and to support the exchange of data of documents and the collaborative use of software applications. Teachers must now exercise their imaginations to make use of this technology.

The next generation Internet project (NGI) is in full swing. Today many universities are communicating with bandwidth a thousand times what you can get from your ISP. It will not take more than a couple of years for these connections to be economically available to most colleges, replacing their current Internet service. It is not too soon to begin to plan for this eventuality, and to begin to experiment with instructional methods and technology that it will make practical and preferred. The ability to synchronize the educational experience of physically dispersed learners will be handed to educators by the pursuit of such technology in today’s industry. Industry has first distributed manufacturing geographically in order to take advantage of local economies, and it is now deeply engaged in developing ways to distribute and coordinate support expertise Higher education must be prepared to adapt its instructional strategies to enable it to ride the coat-tails of industrial telecommunications.